A hybrid knowledge base system for fraud detection using accounting data

Ou Liu, Duanning Zhou

Research output: Unpublished contribution to conferenceUnpublished Conference Paperpeer-review

Abstract

Fraudulence is one of the most popular threats to a company. It is important for a company to detect frauds accurately as soon as possible, so as to protect investors' and customers' interests, and assure the company's revenue. Accounting information systems contain large volume of data that can be analyzed and provide clues for fraud detection. Data mining can be used to analyze users' behavior and patterns, and gives hints to managers about potential frauds. This paper proposes a hybrid knowledge base system for fraud detection, in which ant colony optimization and artificial neural networks are used and combined with experts' knowledge to improve the performance of fraud detection. A system prototype is designed and tested, which illustrates the effectiveness of the approach.

Original languageEnglish
Publication statusPublished - 1 Jan 2014
Event20th Americas Conference on Information Systems, AMCIS 2014 - Savannah, GA, United States
Duration: 7 Aug 20149 Aug 2014

Conference

Conference20th Americas Conference on Information Systems, AMCIS 2014
Country/TerritoryUnited States
CitySavannah, GA
Period7/08/149/08/14

Keywords

  • Accounting information systems
  • Ant colony optimization
  • Fraud detection
  • Knowledge base system

Fingerprint

Dive into the research topics of 'A hybrid knowledge base system for fraud detection using accounting data'. Together they form a unique fingerprint.

Cite this